Re: slowdown in tests

Stas Boukarev <[email protected]> Thu, 5 Feb 2026 18:16:20 +0300
Newsgroups gmane.lisp.steel-bank.devel
Message-ID <CAF63=13p6tvYUj_V5CVSBDthgjzJTX_nEY4w=vB0SBBfLi+MyQ@mail.gmail.com>
That's a way to preserve backwards-compatibility. But as an interface,
that looks very suspicious.

On Thu, Feb 5, 2026 at 6:07 PM Charles Zhang <[email protected]> wrote:
>
> Let’s change :sb-eval to use fast eval with evaluator mode :compile, and make :sb-fasteval use fast eval with default to :interpret.
>
> On Thursday, February 5, 2026, 3:49 PM, Stas Boukarev <[email protected]> wrote:
>
> One qualm I have about swithing to sb-fasteval is that it changes the
> default sb-ext:*evaluator-mode*.
>
> On Thu, Feb 5, 2026 at 5:42 PM Douglas Katzman via Sbcl-devel
> <[email protected]> wrote:
> >
> > I see which change caused it, but I'll note that building with sb-fasteval cut the time down to seconds.
> > I'm inclined to disable the test for sb-eval. But really it's time to make sb-fasteval the default.
> >
> > On Thu, Feb 5, 2026 at 8:08 AM Gábor Melis <[email protected]> wrote:
> >>
> >> Somewhere between e251cf9 and 6db437, genheaders.test.sh slowed down a lot, from about 12-30s to 5+ minutes on my laptop.
> >>
> > _______________________________________________
> > Sbcl-devel mailing list
> > [email protected]
> > https://lists.sourceforge.net/lists/listinfo/sbcl-devel
>
>
>
> _______________________________________________
> Sbcl-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/sbcl-devel


_______________________________________________
Sbcl-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sbcl-devel